You are too close to your own proposal to see its holes. Too invested to be truly critical, too familiar with it to notice the missing piece. Everyone is. It is not a flaw in your thinking; it is a property of being the author. And it happens to be exactly the gap AI is good at filling.
MIRROR, a framework from Dr. Jules White at Vanderbilt University, flips AI from something that produces work into something that interrogates it — holding your own thinking up so you can study it from vantage points you can't occupy on your own. It is one of the methods we teach and apply with teams.
The six checks. Before sending anything important, ask AI to check six things: Missing elements, Interpretations of how the work could be read, Robustness, Risks, Other perspectives, and Refinement. The acronym is the point — it is a repeatable pass, not a vague request for feedback.
The goal isn't a flawless draft. It's clear sight — surfacing the weak points early, while fixing them still costs you almost nothing.
Three depths. There is a ten-minute version for any important email: paste the work in, address the top two or three issues it surfaces, and note any assumptions you cannot yet resolve. There is a thirty-minute version for critical plans and proposals. And there is a weekly team version, where blind-spot detection becomes a standing Best Practice — what are we taking for granted this week, which early warnings deserve attention, and what we would change if we got to begin the week again.
A worked example. "Here is my proposal. What am I missing?" AI returns the unstated assumption, the objection a skeptic will raise, the failure mode no one named. You address the few that matter and send with confidence, having found the gaps while they were still cheap to fix rather than after the decision was made.
Why this matters for a Trusted AI Culture. This is where a Trusted AI Culture compounds. Every important piece of work gets a second set of eyes that never tires and never flatters. It is also the natural close to this series: navigation built the trust, the cost-to-check rule decided what to delegate, MATCH found the work, the Briefing Pattern gave it context, and MIRROR checks the result before it goes out the door.
Build a Trusted AI Culture in your team. Explore the workshop.
Keep reading: Navigation, not generation · Find the work with MATCH
Where does your team stand? Take the AI Blind Spot Assessment — 18 questions, under 3 minutes.
Source framework: Dr. Jules White, Vanderbilt University. Built on his prompt-patterns research; presented here in our own words.
#TrustedAICulture #MIRROR #AIadoption